HuBLE-UK: Lithospheric Experiment Ian Bastow, J-Michael Kendall, George Helffrich, James Wookey, David Thompson SEIS-UK & HuBLE Working Group [email protected] Department of Earth Sciences, University of Bristol, UK PROJECT OVERVIEW SEISMIC STATION CONSTRUCTION RECEIVER FUNCTION ANALYSES We are conducting a broadband seismological experiment in in order to understand better the origins of We deployed 10 broadband seismic stations during summer 2007. 4 sites are housed in secure compounds in As we collect data from the field, we will perform multiple analyses on our seismic data. Receiver function Hudson Bay. Our 10 remote and community based stations complement the broader “POLARIS” network of Community locations: , Cape Dorset, and . 6 stations are in remote analyses (Figure 6) will allow us to analyze velocity discontinuities beneath Hudson Bay. In addition to bulk stations currently operating in the region. Our data will enable us to test hypotheses of root formation and the locations across the northern part of Hudson Bay: (3), , Coats Island, crustal properties (layer thickness and P/S wavespeed ratio) we can investigate the existence of an eclogite persistence of intra-cratonic basins via a range of broadband seismological analyses such as travel-time Mingo Lake and Nottingham Island. While the community stations are run by mains power, our remote sites root beneath the Bay that may be partly responsible for the formation of the basin. tomography, receiver function analyses and SKS shear wave splitting. Some of the major questions we hope to are powered by solar panels; instrumentation are housed in secure steel containers (below). P-to-S conversions generated deeper in the mantle at ~410 and ~660km depth will be deflected up or answer through HuBLE are: down in response to variations in temperature (Figure 7). Thus we will be able to understand better the mantle flow patterns beneath Hudson Bay. • What is the structure and evolution of lithospheric roots and their dynamic interaction with mantle flow? Batteries • How has this intra-cratonic basin formed? Figure 6: Moho structure from • What is the lithospheric structure of the Trans-Hudson Orogen beneath Hudson Bay and the nature of the Satellite modem receiver function analyses. An eclogite root at the base of the GPS antenna Nastapoka Arc? Hudson Bay crust will result in a more • What is the nature of post glacial re-bound in Hudson Bay, and what is its effects on seismicity in this gradational velocity profile than when continental interior region? normal mantle peridotite directly underlies a granitoid crust. Receiver Recording unit function analyses are sensitive to Data will be available to us later this year after the first station service run. Satellite modems at our stations Power regulator these velocity structures - more allow us to monitor the state of health of our stations. After a short period (~3-4 weeks) during the shortest impulsive P-to-S conversions are winter days when we lost power at all stations, the network now appears to be operating well. predicted in the no-eclogite case. After Zandt and Gilbert (2007). Solar panels Figure 7: Investigating the mantle transition zone using receiver Hudson Bay function analyses. (A) In the high Seismometer temperature environment, the 410km discontinuity is depressed and the 660 uplifted, resulting in a thinner transition zone. The opposite effect can be seen in the low temperature region. (B) P-to-S conversions at the transition zone. Modified from Lebedev et al., (2002).

SEISMIC TOMOGRAPHY With a network aperture of ~2000km we will be able to image mantle seismic structures to depths below the transition zone using seismic travel-time tomography. P- and S-wave images will offer lateral resolution of

variations in the nature of the lithospheric root and the flow of mantle material around it. VP/VS ratios will Figure 2: S-wave velocity perturbations provide a means of distinguishing between compositional and thermal effects. Figure 1: Gravity images from the GRACE at 200 km depth from global model 0 (A) project. Note the long wavelength anomaly SAW24B16 (Megnin & Romanowicz, (83°N, 51°W) that characterizes the Hudson Bay region. 2000). The 3% contour outlines the ) 100

(83°N, 110°W) m www.csr.utexas.edu/grace/gallery/gravity k (

approximate region of the tectospheric h t

root beneath Laurentia. p 200 e Craton D Asthenosphere Asthenosphere ? Eroded 300 ? INTRODUCTION AND SCIENTIFIC RATIONALE Figure 3 (above): HuBLE-UK remote ? Despite being one of the most striking features of the continent, the reason for the station construction. 20-40W solar panels (B) 0 on a steel frame re-charge 3×100Ah

existence of Hudson Bay is obscure. It lies in the Precambrian core of North America, which is batteries that power the seismometer and ) 100 m k ( recording equipment. The GPS antenna comprised of the Canadian Shield and contiguous platform regions (Laurentia: Hoffman, 1988). h t Craton

p 200 provides continuous accurate timing e ? Intact ?

D Asthenosphere Asthenosphere The region is underlain by the largest continental root on Earth and is the site of one of the largest information for our data. We communicate negative geoid anomalies (Figure 1); it is also characterized by a broad high-velocity zone with the stations remotely via the satellite 300 ? illuminated in global tomographic models (Figure 2). Only part of the gravity anomaly can be modems that are scheduled to operate twice weekly. attributed to post-glacial isostatic rebound, however, and lithospheric instabilities (Houseman et al., 2000), plume-related lithospheric densification (Kaminski & Jaupart, 2000), the existence of Figure 4 (left): Broadband seismic an eclogite lower crust (Baird et al., 1995) and extension (Roksandic, 1987) are amongst the recording stations throughout the Hudson Figure 8 (above): Vertical cross- various mechanisms proposed to explain the field observations in Hudson Bay. Bay area. HuBLE-UK stations are also sections through end-member shown in the inset figure. models for a craton. A: Lithosphere of the Archean craton has been eroded The Trans-Hudson Orogen, a vast paleo-Proterozoic orogenic belt, played an important role in the and replaced by low density asthenosphere. B: Craton is intact and assembly of Laurentia. It is characterized by extreme salient-reentrant geometry possibly has a lithospheric thickness of more analogous to the western syntaxis of the Himalayan front, but its origins are unknown. Its than 200km. Modified after Ritsema et geometry may reflect a primary shape of the Superior craton and its strong mantle root, but the al., (1998). (45°N, 51°W) current shape of the keel is not well resolved. In the same vicinity, the SE corner of Hudson Bay exhibits a nearly semi-circular coastline known as the Nastapoka Arc. Hypotheses for its REMOTE STATION MODEM CONNECTIONS Figure 9: Knot locations for taut spline parametrization. We The satellite modem facility installed at our remote sites has proved invaluable so far. After all parametrize P- and S-wave slowness using B-splines under tension formation range from meteorite impacts to Archean basement fabrics so we also seek 1800km over a dense grid of knots (Cline 1981). Interpolation between observational constraints here via our experiment. stations powered up after New Year power-downs due to short day-times, two stations slowness values at each knot allows the generation of smooth experienced seismometer and recording equipment problems. These were rectified remotely velocity models through which ray tracing can be performed. The from our UK base-station resulting in us obtaining ~10 months of extra data. Once-weekly equilateral grid consists of 31 knots in depth between 0-1800km, 90 ACKNOWLEDGEMENTS (45°N, 110°W) knots in latitude between 45-83°N and 82 knots in longitude between 51-110°W for a total of 22870 knots parametrizing velocity SEIS-UK have provided invaluable assistance with our field deployment and subsequent remote access of remote sites. The HuBLE-UK monitoring of station power, GPS status and seismometer stability also allow us to prioritise structure. project was funded by NERC grant NE/D012317/1. Logistical support from the CNGO and GSC is also gratefully acknowledged. our efforts at service runs.